{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:VZCQPYOS3OJJ4QCBAQJJWXWAU5","short_pith_number":"pith:VZCQPYOS","schema_version":"1.0","canonical_sha256":"ae4507e1d2db929e404104129b5ec0a74a434b0e9c31d65a523abcb71967d817","source":{"kind":"arxiv","id":"1806.03791","version":1},"attestation_state":"computed","paper":{"title":"The Effect of Network Width on the Performance of Large-batch Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC","cs.LG","math.OC","stat.CO"],"primary_cat":"stat.ML","authors_text":"Dimitris Papailiopoulos, Hongyi Wang, Jinman Zhao, Lingjiao Chen, Paraschos Koutris","submitted_at":"2018-06-11T03:29:17Z","abstract_excerpt":"Distributed implementations of mini-batch stochastic gradient descent (SGD) suffer from communication overheads, attributed to the high frequency of gradient updates inherent in small-batch training. Training with large batches can reduce these overheads; however, large batches can affect the convergence properties and generalization performance of SGD. In this work, we take a first step towards analyzing how the structure (width and depth) of a neural network affects the performance of large-batch training. We present new theoretical results which suggest that--for a fixed number of parameter"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"1806.03791","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2018-06-11T03:29:17Z","cross_cats_sorted":["cs.DC","cs.LG","math.OC","stat.CO"],"title_canon_sha256":"71d813d21b02d0983b8ffa2febe0ab5b2cfdf192adb14d520143894641d7237f","abstract_canon_sha256":"344691da641aae7be143934b93f5538bec0eb85549114718d46491164bba9bee"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:13:42.045937Z","signature_b64":"bxjOGiKHysDwpKa6xJQWTFeds5bstLPIu1SJmwjlS1ydgWRgbyPpihej9ril2eJ1YvTGiWGKja3kULn8vlsSBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ae4507e1d2db929e404104129b5ec0a74a434b0e9c31d65a523abcb71967d817","last_reissued_at":"2026-05-18T00:13:42.045090Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:13:42.045090Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Effect of Network Width on the Performance of Large-batch Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC","cs.LG","math.OC","stat.CO"],"primary_cat":"stat.ML","authors_text":"Dimitris Papailiopoulos, Hongyi Wang, Jinman Zhao, Lingjiao Chen, Paraschos Koutris","submitted_at":"2018-06-11T03:29:17Z","abstract_excerpt":"Distributed implementations of mini-batch stochastic gradient descent (SGD) suffer from communication overheads, attributed to the high frequency of gradient updates inherent in small-batch training. Training with large batches can reduce these overheads; however, large batches can affect the convergence properties and generalization performance of SGD. In this work, we take a first step towards analyzing how the structure (width and depth) of a neural network affects the performance of large-batch training. We present new theoretical results which suggest that--for a fixed number of parameter"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1806.03791","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"1806.03791","created_at":"2026-05-18T00:13:42.045241+00:00"},{"alias_kind":"arxiv_version","alias_value":"1806.03791v1","created_at":"2026-05-18T00:13:42.045241+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1806.03791","created_at":"2026-05-18T00:13:42.045241+00:00"},{"alias_kind":"pith_short_12","alias_value":"VZCQPYOS3OJJ","created_at":"2026-05-18T12:32:59.047623+00:00"},{"alias_kind":"pith_short_16","alias_value":"VZCQPYOS3OJJ4QCB","created_at":"2026-05-18T12:32:59.047623+00:00"},{"alias_kind":"pith_short_8","alias_value":"VZCQPYOS","created_at":"2026-05-18T12:32:59.047623+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2102.01293","citing_title":"Scaling Laws for Transfer","ref_index":136,"is_internal_anchor":true},{"citing_arxiv_id":"2112.00861","citing_title":"A General Language Assistant as a Laboratory for Alignment","ref_index":178,"is_internal_anchor":false},{"citing_arxiv_id":"2207.05221","citing_title":"Language Models (Mostly) Know What They Know","ref_index":255,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VZCQPYOS3OJJ4QCBAQJJWXWAU5","json":"https://pith.science/pith/VZCQPYOS3OJJ4QCBAQJJWXWAU5.json","graph_json":"https://pith.science/api/pith-number/VZCQPYOS3OJJ4QCBAQJJWXWAU5/graph.json","events_json":"https://pith.science/api/pith-number/VZCQPYOS3OJJ4QCBAQJJWXWAU5/events.json","paper":"https://pith.science/paper/VZCQPYOS"},"agent_actions":{"view_html":"https://pith.science/pith/VZCQPYOS3OJJ4QCBAQJJWXWAU5","download_json":"https://pith.science/pith/VZCQPYOS3OJJ4QCBAQJJWXWAU5.json","view_paper":"https://pith.science/paper/VZCQPYOS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1806.03791&json=true","fetch_graph":"https://pith.science/api/pith-number/VZCQPYOS3OJJ4QCBAQJJWXWAU5/graph.json","fetch_events":"https://pith.science/api/pith-number/VZCQPYOS3OJJ4QCBAQJJWXWAU5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VZCQPYOS3OJJ4QCBAQJJWXWAU5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VZCQPYOS3OJJ4QCBAQJJWXWAU5/action/storage_attestation","attest_author":"https://pith.science/pith/VZCQPYOS3OJJ4QCBAQJJWXWAU5/action/author_attestation","sign_citation":"https://pith.science/pith/VZCQPYOS3OJJ4QCBAQJJWXWAU5/action/citation_signature","submit_replication":"https://pith.science/pith/VZCQPYOS3OJJ4QCBAQJJWXWAU5/action/replication_record"}},"created_at":"2026-05-18T00:13:42.045241+00:00","updated_at":"2026-05-18T00:13:42.045241+00:00"}